Nitrogen line spectroscopy in O-stars

This work aims at enabling quantitative optical nitrogen line
spectroscopy in O-stars, in order to improve our knowledge about
these objects, particularly their earliest spectral subtypes. To
this end, nitrogen has proven to be a key element, both in terms of
its potential to infer effective temperatures, and for being the
best tracer for testing the effects of rotational mixing in massive
stellar models, allowing us to further constrain the evolution of
massive stars. To accomplish this study, we used the NLTE (non
local thermodynamic equilibrium) atmosphere/spectrum synthesis code
FASTWIND, and developed a new nitrogen model atom, partly based on
older work, comprising the ionization stages NII to NV. Moreover,
we incorporated the dielectronic recombination mechanism into
FASTWIND, which was previously unable to deal with this process. We
performed an extensive investigation on the line-formation process
of strategic nitrogen emission lines such as NIII4640 and NIV4058,
accounting for a complete treatment of line blocking/blanketing
effects and the presence of a wind, as it is possible when using a
state-of-the-art atmospheric code. Contrasted to the results from
the seminal work on the NIII triplet formation performed by Mihalas
& Hummer (1973), based on much simpler model atmospheres, our
study implies that dielectronic recombination plays only a
secondary role under Galactic conditions. Rather, the emission is
controlled by the stellar wind (Swings mechanism), as long as the
wind is powerful enough to enable a significantly accelerating
velocity field already in the photospheric formation region. For
later spectral O-subtypes, the strength of the emission might be
also affected by an OIII resonance line overlapping with the NIII
resonance line in the EUV. Concerning the emission at NIV4058, we
suggest a rather similar mechanism. Also in this case, the
dominating process is the strong depopulation of the lower level of
the transition, which increases as a function of the wind-strength.
Unlike the NIII triplet emission, however, resonance lines do not
play a role for typical mass-loss rates and below. Using the
updated version of FASTWIND and our new model atom, we derived
nitrogen abundances for a substantial O-star sample in the Large
Magellanic Cloud (LMC). Stellar and wind parameters of our sample
stars were determined by line profile fitting of hydrogen, helium
and nitrogen lines, exploiting the corresponding ionization
equilibria. The bulk of our sample stars turned out to be strongly
nitrogen-enriched, and a clear correlation of nitrogen and helium
enrichment was found. By comparing the nitrogen abundances as a
function of projected rotation velocity with tailored evolutionary
calculations, we identified a considerable number of highly
enriched, but slowly rotating object, which should not exist
according to standard theory of rotational mixing. Our findings
support the basic outcome of previous B-star studies within the
VLT-FLAMES survey on massive stars which pointed to similar
discrepancies. The detection of strong nitrogen enrichment in the
bulk of our sample stars indicates that efficient mixing takes
place already during the very early phases of stellar evolution of
LMC O-stars. In the last part of this thesis, we concentrated on
the applicability of the Walborn et al. (2002) classification
scheme for very early O-stars, which is primarily based on the
relative strengths of the NIV4058 and NIII4640 emission lines. This
scheme has already been used in a variety of studies, but is still
disputed for various reasons. We provided first theoretical
predictions on the NIV4058/NIII4640 emission line ratio in
dependence of different parameters, and confronted these
predictions with results from an analysis of a sample of early-type
LMC/SMC O-stars. Though we found a monotonic relationship between
the NIV/NIII emission line ratio and the effective temperature, all
other parameters being equal, our predictions indicate additional
dependencies on other important stellar parameters, most
significantly, the nitrogen abundance. The relation between the
observed NIV/NIII emission line ratio and the effective
temperature, for a given luminosity class, turned out to be quite
monotonic for our sample stars, and to be fairly consistent with
our model predictions. The scatter within a spectral subtype is
mainly produced by abundance effects. Our findings suggest that the
Walborn et al. classification scheme is able to provide a
meaningful relation between spectral type and effective
temperature, as long as it is possible to discriminate for the
luminosity class.